Patent US5285474A successfully refreshed from Google Patents.
US5285474A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for equalizing a multicarrier signal in a multicarrier communication system
Assignee: Silicon Valley Bank Inc
•
Filing: 1992-06-12
•
Granted: 1994-02-08
•
Expires: 2012-06-12
Abstract
Abstract A method for optimizing a set of equalizer parameters of an equalizer to be used to equalize a multicarrier data signal that has been transmitted through a distorting channel, comprised of the steps of initializing the parameters, repeatedly sending a training sequence through the channel to the receiver, using the equalizer parameters, the received sequence, and a local replica of the training sequence to update a set of channel target response parameters, windowing the channel target response parameters, using the channel target response parameters, the received sequence and the local replica to update the equalizer parameters, and windowing the equalizer parameters. This training process is repeated until a predetermined convergence condition is detected.
Inventors
- 👤 Jacky Chow
- 👤 John M. Cioffi
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-27 08:56
CPC Classifications
H; H04; H04L; H04L27/00; H04L27/01; H04L27/26; H04L27/2601; H04L27/2647
Claims Analysis
Total Claims: 16
Claim 1
Independent Claim
450 words
In a data communication system comprising a multicarrier transmitter which encodes and modulates input data onto a plurality of carriers, generates n samples of a transmit signal, and then appends to the front of the block of N samples a cyclic prefix of L samples before transmitting all (N+L) samples, a channel which severely distorts signals input to it by the transmitter, and a multicarrier receiver which includes an equalizer which partially equalizes the distorted signals it receives from the channel, and means for removing the L samples of the cyclic prefix, and then demodulating and decoding the remaining N samples, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer, wherein L, M and N are positive integers, said method comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters W u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (l) repeating steps (c) through (k) until a predetermined convergence condition is determined.
Claim 2
Dependent Claim
450 words
In a data communication system comprising a multicarrier transmitter which encodes and modulates input data onto a plurality of carriers, generates n samples of a transmit signal, and then appends to the front of the block of N samples a cyclic prefix of L samples before transmitting all (N+L) samples, a channel which severely distorts signals input to it by the transmitter, and a multicarrier receiver which includes an equalizer which partially equalizes the distorted signals it receives from the channel, and means for removing the L samples of the cyclic prefix, and then demodulating and decoding the remaining N samples, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer, wherein L, M and N are positive integers, said method comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters W u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (l) repeating steps (c) through (k) until a predetermined convergence condition is determined.
Claim 3
Dependent Claim
177 words
The method of claim 1, in which step (e), the step of updating the frequency-domain channel target response parameters B u , comprises the steps of: (e1) transforming the samples of said received signal y(D) into the frequency domain to produce a frequency domain received signal Y, (e2) transforming the windowed time-domain equalizer parameters w w (D) to frequency domain to produce a set of windowed frequency-domain equalizer parameters W w , (e3) multiplying said frequency-domain received signal Y by said windowed frequency-domain equalizer parameters W w to produce W w Y, (e4) multiplying said local replica X' by said windowed frequency-domain channel target response parameters B w to produce B w X', (e5) subtracting W Y of step (e3) from B;X' of step (e4) to produce a channel target response error E, and (e6) using said channel target response error E, the local replica X' and the windowed frequency-domain channel target response parameters B to correlate E and X' and perform a Least Mean Squares update of the frequency-domain channel target response parameters B u .
Claim 4
Dependent Claim
94 words
The method of claim 1, in which step (e), the step of updating the frequency-domain channel target response parameters B u , comprises the steps of: (e1) performing a time-domain convolution on said received signal y(D) and the windowed time-domain equalizer parameters w w (D) to produce a time-domain equalized response z(D), (e2) transforming the time-domain equalized response to the frequency domain to produce a frequency-domain equalized response Z, and (e3) dividing the frequency-domain equalized response Z by the local replica X' to produce an updated frequency-domain channel target response parameters B u .
Claim 5
Dependent Claim
175 words
The method of claim 1, 2 or 3 in which step (i), the step of updating the frequency-domain equalizer parameters W u , comprises the steps of: (i1) transforming the samples of said received signal y(D) into the frequency domain to produce a frequency domain received signal Y, (i2) transforming the windowed time-domain equalizer parameters w w (D) to frequency domain to produce a set of windowed frequency-domain equalizer parameters W w , (i3) multiplying said frequency-domain received signal Y by said windowed frequency-domain equalizer parameters W w to produce W w Y, (i4) multiplying said local replica X' by the windowed frequency-domain channel target response parameters B w to produce B w X', (i5) subtracting W w Y of step (i3) from B w X' of step (i4) to produce a equalizer error E, and (i6) using said equalizer error E, the frequency-domain received signal Y and the windowed frequency-domain equalizer parameters W w to correlate E and Y and perform a Least Mean Squares update of the frequency-domain equalizer parameters W u .
Claim 6
Dependent Claim
85 words
The method of claim 1, 2 or 3 in which step (i), the step of updating the frequency-domain equalizer parameters W u , comprises the steps of: (i1) multiplying the windowed frequency-domain channel target response B w by the local replica X' to produce B w X', (i2) transforming the received signal y(D) to the frequency domain to produce a frequency-domain received signal Y, and (i3) dividing B w X' by the frequency-domain received signal Y to produce updated frequency-domain equalizer parameters W u .
Claim 7
Independent Claim
192 words
The method of any of claims 1, 2 or 3, wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) scaling each of the samples of b w (D) by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero.
Claim 8
Dependent Claim
187 words
The method of claim 4 wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) scaling each of the samples of b w (D) by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero.
Claim 9
Dependent Claim
187 words
The method of claim 5 wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) scaling each of the samples of b w (D) by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero
Claim 10
Independent Claim
219 words
The method of any of claims 1, 2 or 3, wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) comparing the total energy of said channel target window group to a predetermined energy threshold, whereby when said total energy is less than said energy threshold, each of the samples of b w (D) is scaled by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero
Claim 11
Dependent Claim
214 words
The method of claim 4 wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) comparing the total energy of said channel target window group to a predetermined energy threshold, whereby when said total energy is less than said energy threshold, each of the samples of b w (D) is scaled by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero
Claim 12
Dependent Claim
214 words
The method of claim 5 wherein step (g), the step of windowing said time-domain channel target response parameters b u (D), comprises the steps of: (g1) searching through all the samples of b u (D) to locate a channel target window group of L consecutive samples wherein said channel target window group has the largest total energy, (g2) copying the samples of said channel target window group to form b w (d) and setting all other samples of b w (D) to zero, and (g3) comparing the total energy of said channel target window group to a predetermined energy threshold, whereby when said total energy is less than said energy threshold, each of the samples of b w (D) is scaled by a normalizing factor so that the sum of the squares of the scaled samples equals a predetermined constant; and wherein step (k), the step of windowing said time-domain equalizer parameters w u (D), comprises the steps of: (k1) searching through all the samples of w u (D) to locate an equalizer window group of M consecutive samples wherein said equalizer window group has the largest total energy, (k2) copying the samples of said equalizer window group to form w w (d) and setting all other samples of w w (D) to zero
Claim 13
Dependent Claim
393 words
In a data transmission system including a transmitter, a distorting channel and a receiver, said receiver including an equalizer which processes a received signal before said signal is input to a demodulator, wherein said data transmission system requires that the impulse response of the received signal be shortened to L samples in the equalizer, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters w u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (1) repeating steps (c) through (k) until a predetermined convergence condition is determined.
Claim 14
Dependent Claim
393 words
In a data transmission system including a transmitter, a distorting channel and a receiver, said receiver including an equalizer which processes a received signal before said signal is input to a demodulator, wherein said data transmission system requires that the impulse response of the received signal be shortened to L samples in the equalizer, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters w u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (1) repeating steps (c) through (k) until a predetermined convergence condition is determined.
Claim 15
Dependent Claim
415 words
In a data storage system including a write mechanism, a distorting storage medium, a read mechanism, and a receiver which processes a received signal read from the storage medium by the read mechanism, said receiver including a decoder and an equalizer which processes said received signal before said signal is input to the decoder, wherein said data storage system requires that the impulse response of the received signal be shortened to L samples in the equalizer, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters w u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (l) repeating steps (c) through (k) until a predetermined convergence condition is determined.
Claim 16
Dependent Claim
415 words
In a data storage system including a write mechanism, a distorting storage medium, a read mechanism, and a receiver which processes a received signal read from the storage medium by the read mechanism, said receiver including a decoder and an equalizer which processes said received signal before said signal is input to the decoder, wherein said data storage system requires that the impulse response of the received signal be shortened to L samples in the equalizer, a method of calculating a set of M windowed time-domain equalizer parameters w w (D) of said equalizer comprising the steps of: (a) initializing said windowed time-domain equalizer parameters w w (D) and a set of windowed frequency-domain channel target response parameters B w to some predetermined initial values, (b) calculating, in said transmitter, a set of N frequency domain samples of a repetitive multicarrier signal X, (c) encoding and transmitting said multicarrier signal X from said transmitter through said channel and into said receiver to produce a received signal y(D), (d) generating, in said receiver, a set of N frequency domain samples of a local replica X' of said repetitive multicarrier signal, (e) using said windowed time-domain equalizer parameters w w (D), the samples of said local replica X' and said received signal y(D) to update said frequency-domain channel target response parameters B u , (f) transforming said frequency-domain channel target response parameters B u into the time domain to produce a set of time-domain channel target response parameters b u (D), (g) windowing said time-domain channel target response parameters b u (D) by selecting L consecutive samples of b u (D) to produce a set of windowed time-domain channel target response parameters b w (D), (h) transforming said windowed time-domain channel target response parameters b w (D) into the frequency domain to produce a set of windowed frequency-domain channel target response parameters B w , (i) using said windowed frequency-domain channel target response parameters B w , the samples of said local replica X' and said received signal y(D) to update a set of frequency-domain equalizer parameters W u , (j) transforming said frequency-domain equalizer parameters W u into the time domain to produce a set of time-domain equalizer parameters w u (D), (k) windowing said time-domain equalizer parameters w u (D) by selecting M consecutive samples of w u (D) to produce a set of windowed time-domain equalizer parameters w w (D), and (l) repeating steps (c) through (k) until a predetermined convergence condition is determined.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1997-02-07 | 1997-08-09 | 1998-02-08 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2001-02-07 | 2001-08-09 | 2002-02-08 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2005-02-07 | 2005-08-09 | 2006-02-08 | $7700.00 | Window Closed / Lapsed |